All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
The authors have satisfactorily addressed all the reviewers'comments, including those from reviewer 2, as assessed by the Editor. The manuscript is therefore ready for publication in the present form.
[# PeerJ Staff Note - this decision was reviewed and approved by Celine Gallagher, a PeerJ Section Editor covering this Section #]
The authors have addressed my concerns. I have no further comments.
The authors have addressed my concerns. I have no further comments.
The authors have addressed my concerns. I have no further comments.
Well Done
Well Done
Well Done
Well Done
All the reviewers have expressed appreciation towards this study. The authors are invited to address the issues raised by the reviewers, in particular those related to the Methodology and Discussion.
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
The manuscript is generally written in clear, professional English.
The research question is appropriate and clinically relevant. However, methodological rigour needs strengthening. The literature search was restricted to four English‑language databases. Do so may introduce language bias and may omit pertinent studies. Expanding the search to CENTRAL, Scopus and non‑English databases (or explicitly justifying their exclusion) is advisable.
Study quality was assessed with the Newcastle–Ottawa Scale, but because all included cohorts are non‑randomised, ROBINS‑I provides a more granular evaluation and should replace or complement the current approach.
The decision to select the hazard ratio with the largest effect size when multiple glycaemic‑variability metrics were reported is problematic. It can inflate the pooled estimate. Using a pre‑specified hierarchy or multilevel model would be less biased. Substantial heterogeneity (I² ≈ 92 %) is acknowledged but explored only with leave‑one‑out analyses and categorical subgroup test
While the pooled hazard ratio points toward a harmful association between glycaemic variability and heart failure, several issues limit confidence in this estimate. Residual confounding remains a concern. Key cardioprotective medications (SGLT2 inhibitors, GLP‑1 receptor agonists) were not adjusted uniformly across studies and socioeconomic factors were largely unmeasured.
High heterogeneity further undermines the generalisability of the summary effect. Without robust meta‑regression, it is unclear whether differences in diabetes prevalence, follow‑up duration or variability metrics drive the divergence. Two Chinese cohorts appear to originate from overlapping health‑check populations and must be clarified (unknown if this is double-counting participants)
No comment
no comment
no comment
The article uses good statistical methods and the results are properly described
This meta-analysis on GV and risk of HF is well-written and uses proper methods. The topic is novel and interesting. A few minor points:
-Table 1 is way too busy. I would remove the columns for mean age and male sex for sure, and maybe even the column on variables adjusted; that way the HR result will be more prominent.
-If space allows, mention p value for publication bias in abstract
-I may have missed it, but would it be possible for you to report in the results section the HR for studies that used HBA1c and fasting glucose separately? So that we can compare the strength of these (I did not see it in the main results; if it is already elsewhere, please include it there)
- Language and Clarity: The manuscript is written in clear, professional English. Minor grammatical improvements could enhance readability, particularly in the Introduction and Discussion sections.
- Structure and Formatting: The article follows a standard scientific format (Abstract, Introduction, Methods, Results, Discussion, Conclusion). Figures and tables are relevant and well-labeled.
- Literature Context: The authors provide a comprehensive background, citing recent and relevant studies. The rationale for the meta-analysis is well established.
- Data Availability: Raw data and supplementary materials are appropriately shared.
- Research Question: The study addresses a meaningful clinical question—whether long-term glycemic variability (GV) is associated with heart failure (HF) risk.
- Methodology: The authors conducted a systematic review and meta-analysis following PRISMA guidelines. Inclusion/exclusion criteria are clearly defined using the PICOS framework.
- Reproducibility: Methods are described in sufficient detail to allow replication. The use of multiple GV indices and subgroup analyses adds robustness.
- Statistical Rigor: The analysis uses appropriate random-effects models and sensitivity analyses. Heterogeneity is acknowledged and explored.
- Data Interpretation: Conclusions are well-supported by the data. The authors avoid overstatement and appropriately distinguish correlation from causation.
- Limitations: The authors acknowledge key limitations, including heterogeneity in GV definitions and the observational nature of included studies
- Strengths:
- Large sample size (over 4.2 million participants).
- Comprehensive subgroup and sensitivity analyses.
- Clear clinical relevance for diabetes and cardiovascular care.
- Suggestions for Improvement:
- Consider standardizing GV definitions or discussing implications of variability across studies.
- Expand discussion on potential clinical applications of GV monitoring.
- Clarify whether HF subtypes (e.g., HFpEF vs. HFrEF) were distinguishable in the included studies.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.